Efforts to make the nuclear industry safer are more important than ever in the wake of the Japanese disaster. An enormous amount has been written about how to "create a safety culture," and it usually ends up with a long list of attributes of a culture and very little insight into the fundamental issues that underlie safety in all high hazard industries. I have observed the great energy and effort that goes into this process in my role as a part-time consultant and member of the Advisory Council of INPO.Carrillo's paper is an important addition to this difficult dialogue in highlighting that we may be using the wrong models of how to think about safety in the first place. Her emphasis on complexity theory, sense making, and polarity theory focus us on the two most fundamental problems of safety: (a) We will never be able to predict all the things that can go wrong, that nature will throw at us, that human beings will, in their efforts to do things better, actually make things more complex and, therefore, maybe worse, and; (b) We will never be able to avoid the polarity between absolute safety (at any cost) and competing economic and psychological values. In our own daily behavior we can see how the need to "accomplish things, get to places, do things in a timely and satisfying way, and have fun" tempts us into "risky" behavior, seen most clearly in the way we drive. But we try to avoid what some might call "reckless" behavior, as defined by consensus of others doing the same thing. In high hazard industries recklessness is totally unacceptable but our goal would be to find ways of avoiding even risky behavior so that the public and the employees are kept safe.Effective management in high hazard industries therefore must focus on not only avoiding recklessness but also preparing employees for the unexpected, what has usefully been labeled the "unknowable unknown." If surprises have to be dealt with, we hope that employee innovation will minimize risk and we hope that management can create the conditions and incentives to enable employees to balance or even integrate the polarities implied by safety versus productivity. We should have learned by now that better design or more detailed procedures is only a partial answer. In fact, the more we try to design fail-safe systems and the more we write procedures for how to do things, the greater the complexity and the potential of surprises of all sorts that we hope the operators in our plants have the ingenuity to deal with. Without their ability to make sense out of surprises and innovate to get the job done we would be much worse off. If we take complexity and polarity theory seriously it will point the way out of these dilemmas-we have to keep learning. We have to learn to think in terms of new models and develop some new skills.Becoming skilled learners and sense makers will be the keys to a safer future. In that regard learning how to be more helpful to each other will be the key because learning and sense making is a joint effort that hinges on mutual trust and mutual help. From working in the safety industry, I learned from employees that the biggest obstacle to improving safety performance is failed communication and lack of trust. Employees feel management doesn't listen. Management feels employees don't understand the bigger picture and suffer from entitlement. We can only fix that problem by creating a climate where employees feel they can tell their boss the truth. Bosses need to communicate that they really need employees' help. Managers need to know what is going on. However, employees won't speak up unless they trust management. The trust is developed as they see management respond to the information. Better information leads to better decisions. In my experience this happens when managers are able to ask for help and employees feel their help is needed and valued.How can we construct the relationships and communication structures that create successful and safe organizations? How do you develop trust? What if I want to trust more and I don't know how? What if I trusted someone and they betrayed me? Helping is the basis of trust. Trust is the basis of communication. Communication is the basis of organizational effectiveness. We have to pay more attention to mutual helping to create both safety and effectiveness. I have reached the conclusion that helping skills at all levels of an organization will be the necessary ingredient to a more effective and safe nuclear industry.